coordinate limit forces to limit range of motion?

Ashley Heers · Post by **Ashley Heers** » Mon Apr 23, 2018 1:37 pm

Hi all,

I am trying to add coordinate limit forces (CLFs) to my model of a flapping bird wing, to restrict the range of motion at the shoulder joint (i.e., my custom joint has no restrictions but in reality there is a limited range). For each coordinate (elevation-depression, protraction-retraction, and long axis rotation) I have placed the upper and lower limits at the maximum and minimum values of those coordinates in a live animal. I made the upper and lower stiffness 0.5, set damping to 0.025, and set the transition width to 5. I've tried many different combinations of stiffness and transition width, as well as different upper and lower limits, and I can never get the CLFs to actually help. Best case scenario, during CMC, the CLFs have no effect (i.e., do not change the reserve actuators compared to an identical simulation without CLFs), but most combinations I've tried actually increase the magnitude of the reserve actuators.

Any suggestions as to what I am doing wrong?

Alternatively, are there any other ways to restrict range of motion during CMC? Obviously the bird can't move it's wing 360 deg in every direction, and this is at least partially due to the skeletal geometry of the joint as well as ligaments and other bones. I'm just trying to figure out how to represent this; my reserve actuators are too high (~30-50% of the total torque) and I'm fairly certain its because I can't restrict the range of motion. I've checked /adjusted EVERYTHING else to try and identify the problem (joint location, different kinematics, muscle architecture and moment arms, activation-deactivation dynamics, tendon slack lengths and stiffness, optimal force of reserve actuators, etc), but nothing had any significant effect on the magnitude of reserve actuators. I don't think my problem is with the muscles: the activations, forces, and power outputs of my modeled muscles (as calculated by CMC) are on par with in vivo data.

Any suggestions are most welcome, and thanks in advance!

Ashley